Raw DNA answers
Raw DNA questions: files, privacy, accuracy, and health limits
Start here for the short answer, then follow the evidence and the detailed guide. These answers describe what supported consumer DNA files can and cannot establish; they are informational, not diagnostic or individualized medical advice.
Published by DNA Layer · Product behavior and primary sources reviewed 26 August 2026 · No independent medical review is claimedTwelve direct answers
Each short answer is followed by the evidence boundary, the most important limitation, named sources, and the canonical page for the full explanation. DNA Layer is the publisher and one of the services discussed; read the company facts and evidence methodology.
What can I learn from my raw DNA data?
Direct answer: Raw DNA can support selected findings about traits, medication response, carrier status, and health-related associations when the relevant markers were tested and the evidence is strong enough. It cannot show every variant or predict your future by itself, because lifestyle, environment, family history, ancestry, and untested DNA still matter.
What supports it: A raw consumer file contains genotype calls at positions selected by the testing provider. Those calls can be matched to published evidence, but a useful result depends on correct parsing, allele orientation, genome build, evidence quality, and the marker actually being present.
Important limitation: The file is a partial measurement, not a complete genome or a complete picture of health.
Sources: MedlinePlus Genetics: what direct-to-consumer results mean; NHGRI: direct-to-consumer raw-data interpretation.
See the seven checks behind responsible raw-DNA interpretation →
Can raw 23andMe data reveal health risks?
Direct answer: Yes. A 23andMe raw file contains genotype calls at selected positions, and some positions have health-related evidence. A careful analysis may surface supported risk, carrier, trait, or medication-response context. It is not a complete health screen, and consequential findings may be incomplete, false positive, or require clinical confirmation.
What supports it: 23andMe makes the raw genotype export available separately from its reports. Independent interpretation can match present markers to evidence, while FDA and NHGRI guidance make clear that different consumer tests examine different variants and that important results may need confirmation.
Important limitation: A result cannot establish that an untested variant is absent or that a condition will or will not develop.
Sources: 23andMe: Accessing Your Raw Genetic Data; U.S. FDA: Direct-to-Consumer Tests; NHGRI: direct-to-consumer raw-data interpretation.
See what DNA Layer can add to a 23andMe file →
Can AncestryDNA raw data show health risks?
Direct answer: Some markers in an AncestryDNA raw file can have health-related evidence, so careful third-party analysis may surface relevant associations. The standard AncestryDNA product is focused on ancestry, matching, and traits, and its raw export is not a diagnosis. Consequential findings may be incomplete, false positive, or require clinical confirmation.
What supports it: Ancestry provides a downloadable DNA-data file containing tested genotype calls. A health-analysis service can use supported calls that overlap its evidence panel, but the useful coverage depends on the exact file rather than the provider’s total advertised marker count.
Important limitation: An ancestry-focused genotyping array is not whole-genome sequencing and cannot answer every health question.
Sources: AncestryDNA FAQ: Downloading your DNA Data; MedlinePlus Genetics: what direct-to-consumer results mean; NHGRI: direct-to-consumer raw-data interpretation.
See what DNA Layer can add to an AncestryDNA file →
What DNA file should I upload?
Direct answer: For 23andMe, use the ZIP whose name begins with genome. For AncestryDNA or MyHeritage, use the original DNA-data ZIP downloaded from the provider. Do not upload a PDF report, screenshot, family-tree GEDCOM, ancestry-coordinates file, or a VCF: DNA Layer does not currently accept VCF files.
What supports it: The supported export is a machine-readable genotype table, normally delivered as a ZIP. Reports, screenshots, family trees, and ancestry-coordinate files contain different information and cannot substitute for the marker rows used during deterministic matching.
Important limitation: Provider interfaces and filenames can change, so use the current provider-specific download instructions.
Sources: 23andMe: Accessing Your Raw Genetic Data; AncestryDNA FAQ: Downloading your DNA Data.
Choose your provider and find the correct file →
Does DNA Layer store my DNA?
Direct answer: DNA Layer reads the original raw file locally in your browser and does not retain that file. It sends only the supported marker derivatives needed to build your layer and privately stores the resulting derived findings so you can return to them. Derived findings remain sensitive even though the original file is not stored.
What supports it: The production upload flow filters the source file in the browser before sending the supported marker data and technical metadata used to build the report. Account-scoped findings remain available until the user deletes them.
Important limitation: Local raw-file handling reduces the data transferred; it does not make derived genetic findings anonymous or risk-free.
Sources: DNA Layer privacy policy.
Read the full DNA Layer data-flow explanation →
Is it safe to upload raw DNA data?
Direct answer: No DNA upload is risk-free. Before using any service, check whether the whole file leaves your device, what raw or derived data is retained, who can access it, whether secondary use is allowed, and how deletion works. Prefer the minimum data flow needed for the result you want.
What supports it: The FTC and MedlinePlus both advise consumers to inspect access, use, safeguards, ownership changes, and deletion. Genetic data is distinctive and can also reveal information relevant to biological relatives, so a generic promise of privacy is not enough.
Important limitation: A strong privacy design does not by itself prove that an interpretation is accurate or clinically valid.
Sources: U.S. FTC: Keep people’s sensitive DNA information private; MedlinePlus Genetics: limitations and privacy questions.
Use the five-question private-analysis checklist →
How accurate are consumer DNA files?
Direct answer: Accuracy has several layers: the provider must call the genotype correctly, the tool must parse the file and allele orientation correctly, and the interpretation must match current evidence without overclaiming. Consumer arrays test selected positions rather than the whole genome, so important findings can still be absent, incomplete, or false positive.
What supports it: A technically correct row can still be clinically misleading if the genome build, strand, or evidence context is wrong. DNA Layer performs deterministic marker matching before explanation and keeps evidence strength and limitations attached to supported findings.
Important limitation: Important consumer findings may require confirmation with an appropriate clinical test before healthcare decisions.
Sources: NHGRI: direct-to-consumer raw-data interpretation; U.S. FDA: Direct-to-Consumer Tests; DNA Layer evidence methodology.
See how format, alleles, evidence, and confirmation are checked →
Can raw DNA data diagnose a condition?
Direct answer: No. Consumer raw data and third-party interpretations are informational, not diagnostic. Diagnosis depends on the clinical question, the right test, confirmation where appropriate, symptoms, family and medical history, and professional interpretation. Do not start, stop, or change medication, screening, or treatment from a consumer raw-data result alone.
What supports it: MedlinePlus says direct-to-consumer results generally need additional testing ordered through a healthcare provider before they can be considered diagnostic or used for medical decisions. FDA guidance likewise distinguishes consumer tests by what each test actually examines.
Important limitation: A negative or missing consumer result cannot rule out a condition.
Sources: MedlinePlus Genetics: limitations and privacy questions; U.S. FDA: Direct-to-Consumer Tests.
See how DNA Layer keeps evidence and limitations visible →
Does a missing marker mean I do not have the variant?
Direct answer: No. A marker can be missing because the provider’s array did not test that position, the file did not report it, or the call did not pass quality checks. Absence from a consumer file is not proof of biological absence, and DNA Layer does not invent a result for an unsupported position.
What supports it: Genotyping arrays read a predetermined set of locations. Different chip versions and providers include different positions, so the same evidence panel can have different usable coverage across two otherwise valid files.
Important limitation: Only a test designed to answer the clinical question can establish whether a consequential variant is present or absent.
Sources: NHGRI: direct-to-consumer raw-data interpretation; U.S. FDA: Direct-to-Consumer Tests.
Compare 23andMe and AncestryDNA coverage without using marker-count shortcuts →
What is the difference between genotyping and sequencing?
Direct answer: Genotyping arrays read a fixed list of already-known DNA positions. Whole-exome sequencing reads much of the protein-coding portion, while whole-genome sequencing reads a much larger share of the genome base by base. Sequencing can cover variants an array never tested, but coverage, quality, interpretation, and clinical purpose still matter.
What supports it: 23andMe-, AncestryDNA-, and many MyHeritage-style exports are array-based genotype tables. WES and WGS commonly produce VCF files and can support different questions because they measure a broader set of positions directly.
Important limitation: DNA Layer currently supports the listed consumer raw-data exports, not WES or WGS VCF files.
Sources: MedlinePlus Genetics: what direct-to-consumer results mean; DNA Layer evidence methodology.
Read the full genotyping-versus-sequencing explanation →
Do I need a new DNA test to use DNA Layer?
Direct answer: No, if you already have a supported raw-data export from 23andMe, AncestryDNA, MyHeritage, or TellMeGen. DNA Layer uses that existing file to build a free personalized preview. A new test may only be relevant if your current file is unavailable, unsupported, or cannot cover the question you need answered.
What supports it: Your inherited DNA sequence is a durable baseline, while the evidence used to interpret supported markers can improve over time. The practical requirement is the original machine-readable export, not a new saliva sample.
Important limitation: An older consumer array still cannot provide positions it never measured, regardless of how current the interpretation is.
Sources: DNA Layer evidence methodology; 23andMe: Accessing Your Raw Genetic Data; AncestryDNA FAQ: Downloading your DNA Data.
See what to do with an older DNA test →
Can ChatGPT or Claude analyze a raw DNA file?
Direct answer: They may be able to open a text-based export, but opening rows is not the same as reliably identifying the provider format, normalizing variants, matching a controlled evidence set, or applying clinical safety rules. Check current privacy settings before sharing genetic data, and prefer giving an assistant a verified, cited report instead.
What supports it: Both products support file uploads, with retention and model-improvement handling that varies by product and settings. DNA Layer instead parses and matches supported markers deterministically, then produces a cited report that can be used for follow-up questions.
Important limitation: No privacy setting makes sharing genetic information risk-free, and a general chatbot should not be treated as a diagnostic service.
Sources: OpenAI: File Uploads FAQ; Anthropic: sensitive data in consumer Claude accounts; DNA Layer evidence methodology.
Compare the raw-file and verified-report workflows →